This column is a look back at the week that was in AI. Read the previous one here.

As the generative AI race gets hotter and hotter, it seems like Big Tech is begging the government to get more involved.

Just a couple weeks ago we discussed Microsoft’s machinations to try to avoid more antitrust scrutiny, even as it announces deal after deal in the AI space. This includes its very unusual relationship with generative AI startup Inflection AI which is now being referred for an initial merger investigation in the U.K.

Well, there was more this week — this time from good friend Google.

The search titan came to a “non-exclusive” licensing agreement with AI chatbot startup Character.ai for its LLM technology, and also brought the company’s co-founders back to Google — where they were before starting Character.ai.

The strange part of the deal is it is being reported that Character’s staff was told investors would be bought out at a $2.5 billion valuation — a nice uptick from the $1 billion the company was valued at after closing a $150 million Series A led by Andreessen Horowitz.

The odd part isn’t the valuation — it’s the fact this is being structured as anything but an acquisition.

Investors are being bought out, a licensing agreement for the satrtup’s technology is in place, and co-founders are leaving to join the licensee — but it’s not an acquisition?

AI has seemingly made venture capitalists lose some touch with reality when it comes to valuations, and apparently it is making corporations get very, shall we say, “creative” in how they structure deals.

It seems reasonable to expect this deal to garner the same amount of scrutiny the Microsoft/Inflection deal did, regardless of how anyone is trying to frame the optics.

Of course, Google has bigger fish to fry now that it has been found to have acted illegally to maintain a monopoly in online search. Nevertheless, the search giant may soon find a new wave of regulators looking at it closely.

Things that caught our eye and other stuff:

• We knew it couldn’t be that easy. Earlier this year, Elon Musk filed a lawsuit against OpenAI and its executives — including co-founder and CEO Sam Altman — for allegedly breaching the founding agreement of the company to develop artificial general intelligence for the benefit of humanity. Just a few months later, Musk unexpectedly dropped the lawsuit. Now — he’s back again. In the revived complaint, the mercurial Musk says he was manipulated into believing OpenAI would be a nonprofit and he was deceived in helping fund the company and attract talent. Like most things with Musk, just give this two or three months, and it likely will pass.
• Investors continue to put money into Chinese OpenAI competitors as the country tries not to fall behind in the generative AI race. Moonshot AI has closed on a fresh $300 million from the likes of Tencent Holdings at a valuation of $3.3 billion, per Bloomberg. Earlier this year it was reported China’s artificial intelligence startup Moonshot AI raised more than $1 billion in a funding round led by Alibaba Group Holding and HongShan, formerly Sequoia Capital China. Just last month, Tencent and Alibaba contributed to China-based generative AI startup Baichuan Intelligence’s $700 million round.
• New York-based Thrive Capital raised $5 billion across two new funds to look at new AI opportunities, per WSJ. The funds include a $4 billion growth fund and $1 billion for early-stage investments. The firm, co-founded by Josh Kushner, has recently made big bets on film studio A24 and cybersecurity firm Wiz

Related reading:

• Eye On AI: Microsoft Continues To Try To Tiptoe Around Regulation

Illustration: Dom Guzman